Learning, Prefrontal Cortex, and Multiple Memory Systems Learning, to be useful, must be sensitive to the relevant features of situations, guided by environmental opportunities, and informed by past actions in similar circumstances. In other words, learning is guided by memory. The neuronal mechanisms that integrate learning and memory are largely unknown. The hippocampus is crucial for learning facts and remembering events, the neostriatum is important for habit learning, and the prefrontal cortex (PFC) is needed to modify previously learned responses flexibly. Dysfunction of each of these brain regions or their disconnection contributes to neuropsychiatric disorders including dementia, PTSD, and schizophrenia. This proposal will investigate how these structures interact during memory-guided learning. The experiments, part of a larger research program on how prefrontal cortex contributes to memory and cognition, will test the hypothesis that interactions between PFC, hippocampal and dorsolateral striatal (DLS) circuits provide key mechanisms for memory guided learning by integrating abstract rules, event sequences, and stimulus-directed actions. The specific aims will investigate these mechanisms by combining behavior analysis, temporary inactivation, simultaneous recording of neuronal activity, and deep brain stimulation. Aim 1 will assess the functional interactions between the PFC, hippocampus, and DLS the during learning by temporary disruption of local circuits. Rats will be trained to two behaviorally identical + maze tasks, one that requires the hippocampus, the other the DLS for initial learning~ the PFC is needed to switch between them. Interactions between the PFC and the other structures will be tested by temporarily the mPFC and one of the other structures both the on opposite side of the brain. If PFC interactions are required for flexible learning, then the crossed inactivation should produce asymmetric impairments in switching from one strategy to the other. Aim 2 will record neuronal activity in the three structures simultaneously to determine how activity within and between the PFC and the other structures predict learning. We recently identified EEG patterns in the hippocampus that predicted memory retrieval, and discovered that DBS could both mimic these patterns and restore memory in otherwise amnestic animals. Aim 3 will therefore test the causal relationships between the PFC and the other structures by combining unilateral inactivation, bilateral recordings, and DBS. Recording in PFC while disrupting activity unilaterally in the hippocampus or DLS, or vice versa, will determine the extent to which normal coding in each structure depends on the other, and how these interactions influence learning. Targetted patterns of DBS will be used to mimic identified signals within and between circuits to determine if the effects of inactivation can be overcome, or learning strategy modified. The outcome will advance neuroscience by revealing how the PFC, hippocampus, and DLS interact to allow memory-guided learning, and will inform emerging treatments for neuropsychiatric disorders that involve disintegration of prefrontal cortex, hippocampal, and striatal functions, including schizophrenia and Alzheimer's disease.